# vercel ai sdk Open-source TypeScript toolkit for building AI-powered applications and agents with multi-provider model support and streaming UI capabilities. This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles. ## AI SDK harness layer now supports native subscription authentication DevFeed: [AI SDK harness layer now supports native subscription authentication]() Original publisher: [Read original article]() Author: Felix Arntz Published: 2026-09-14T21:28:00Z Content type: release Language: en Sources: [Vercel News]() Topics: [vercel ai sdk](), [Authentication](), [OAuth](), [codex](), [Claude Code]() Tags: [ai](), [authentication](), [claude-code](), [codex](), [coding-agents](), [copilot](), [cursor](), [gateway](), [oauth](), [opencode](), [sdk]() ### AI overview The AI SDK harness layer now supports authenticating supported coding-agent harnesses through their native subscriptions. It uses explicit provider credentials when available, otherwise falls back to a host-native subscription without requiring code or configuration changes. ### Source excerpt The AI SDK harness layer now supports authenticating harnesses through their native subscriptions, where the underlying harness supports them. The harness layer runs different coding agents through the same HarnessAgent interface, so you can switch agents without changing your application code. No code changes or new settings are required. The direct authentication mode uses explicit provider environment credentials when they are present, and otherwise a native subscription found on the host. The default auto mode does the same when no AI Gateway credentials are set. The ai-gateway mode never reads native subscriptions. Native subscription credentials stay on the host. Just like with API keys, credentials are resolved at the host boundary. OAuth access tokens are refreshed there as needed. Where the sandbox supports it, the harness receives placeholder credentials and the real token is injected into outbound requests on the host. Native subscriptions work across harness adapters wherever the harness itself supports subscription login. These include Claude Code, Cline, Codex, Cursor, fx, GitHub Copilot, Grok Build, OpenCode, and Pi. Read the harness documentation to learn more. Read more ## The Grafana AI SDK for Go: a shared foundation for building AI applications DevFeed: [The Grafana AI SDK for Go: a shared foundation for building AI applications]() Original publisher: [Read original article]() Author: Luccas Quadros Published: 2026-09-12T11:22:06.456390Z Content type: article Language: en Sources: [Grafana Labs blog on Grafana Labs]() Topics: [vercel ai sdk](), [SDKs](), [Go Language](), [AI Development](), [Large Language Model](), [agent observability](), [Model Context Protocol](), [React]() Tags: [agent-observability](), [ai](), [analytics](), [backend](), [building](), [go](), [grafana](), [logs](), [metrics](), [monitoring](), [observability](), [sdk](), [streaming](), [tools](), [traces]() ### AI overview Grafana Labs introduces the Grafana AI SDK for Go, an open-source shared foundation for building AI applications. The SDK standardizes model calls, streaming, tool execution, structured output, multi-step agents, workflow controls, and operational features such as retries, logging, metrics, and Agent Observability. It also supports streaming Go backends to Vercel AI SDK frontend hooks. ### Source excerpt Starting an experiment with an LLM has never been easier. Keeping a growing collection of those experiments consistent is another matter. Earlier this year, as more teams began exploring AI features here at Grafana Labs, we repeatedly encountered the same pattern: a new experiment would start, move quickly, and build its own client for whichever model provider it needed. The next experiment would do the same, with a slightly different abstraction for streaming, tools, errors, or provider configuration. This was understandable, given the circumstances. Model providers were changing quickly, our teams were learning quickly, and coding agents made it possible to turn an idea into a working integration faster than ever. But that speed also made it easier for every integration to develop its own architecture. Eventually, we were maintaining a collection of solutions to what was essentially the same problem. And since most of our backend is written in Go, we built the Grafana AI SDK for Go to give our teams a shared foundation to work from. It provides common interfaces for calling models, streaming responses, executing tools, producing structured output, and running multi-step agents. It also speaks the protocol used by Vercel AI SDK frontend hooks, so a Go backend can stream directly to useChat, useCompletion, and useObject. We built it because we needed it inside Grafana Labs, but we open sourced it last month (alongside a broader collection of tools we released for building, operating, and understanding AI systems during our first Grafana Labs AI Week) because we think other teams building AI applications in Go are likely to encounter many of the same problems. We would like to build the next part together, so in this blog I'll tell you a bit more about the project, including how you can put it to use today, as well as how you can help us improve it. What teams can build with it today The SDK supports both simple model calls and larger application workflows: Generate ## How to Build an AI Chat App Interface With the Vercel AI SDK and Shadcn/ui DevFeed: [How to Build an AI Chat App Interface With the Vercel AI SDK and Shadcn/ui]() Original publisher: [Read original article]() Author: Vaibhav Gupta Published: 2026-09-11T16:21:29Z Content type: tutorial Language: en Sources: [freeCodeCamp Programming Tutorials: Python, JavaScript, Git & More]() Topics: [vercel ai sdk](), [AI Chat](), [SDKs](), [Node.js]() Tags: [ai](), [app](), [how-to](), [llm](), [mcp](), [next-js](), [node](), [openai](), [react](), [shadcn-ai-chat-app](), [shadcn-ui](), [streaming](), [tailwind](), [typescript](), [ui](), [vercel](), [vercel-ai-sdk]() ### AI overview Tutorial for building a streaming AI chat interface in Next.js with the Vercel AI SDK and shadcn/ui. ### Source excerpt Every other AI product you open today has the same screen: a message list, a text box at the bottom, and words that stream in one token at a time. It looks simple, but it's not simple to build well. Y ## How Featured's users make 100K media pitches per month on Vercel DevFeed: [How Featured's users make 100K media pitches per month on Vercel]() Original publisher: [Read original article]() Author: Susan Aziz Published: 2026-09-11T04:00:00Z Content type: article Language: en Sources: [Vercel News]() Topics: [migration](), [vercel ai sdk](), [AI Chat](), [SDKs]() Tags: [ai](), [ai-gateway](), [aws](), [featured](), [infrastructure](), [migration](), [models](), [vercel](), [vercel-ai-sdk](), [workflow]() ### AI overview Featured describes migrating 374 Sanity sites from AWS Elastic Beanstalk to Vercel while using Vercel's AI SDK, AI Gateway, and Workflow SDK to reduce infrastructure work for a three-engineer team. ### Source excerpt Featured on Vercel 3 engineers supporting 3 brands and 100,000+ users on Vercel Migrated 374 Sanity sites from AWS Elastic Beanstalk to Vercel AI SDK and AI Gateway power Featured's chat bot across 17 models Workflow SDK replaced custom long-running job infrastructure Featured is a co-pilot for public relations (PR) that subject matter experts and PR teams use to find media opportunities. Tell Featured's agents what you know, and it surfaces opportunities across journalist requests, podcasts, awards, and GEO, with no PR background required. Founder Brett Farmiloe knows from experience how hard and time consuming getting press is. He spent 10 years running Markitors, a digital marketing agency with 500 small business clients. Every client, from an eyelash extension supplier to an equipment financing company, had real expertise to share, but no way to get it in front of journalists. PR, as Farmiloe puts it, "has always been about who has access to what." He founded Featured to change the question from who has access to who has knowledge. Featured connects one of their users with a journalist or publisher every 6 seconds. Their agents deliver more than 100,000 media pitches per month, and have sent over 100 million Help A Reporter Out (HARO) emails in the past year. Behind it all is an engineering team of just three people. With a team that lean, there's no time to manage servers or piece together custom integrations. Every hour spent on infrastructure is an hour taken away from building features what will help their customers land more media placements. The cost of managing infrastructure by hand Before Vercel, Featured's infrastructure work pulled the team away from product development. Hosting lived on AWS Elastic Beanstalk, AI features depended on custom provider integrations, and long-running, multi-step jobs ran on separate orchestration infrastructure. Each layer worked, but each one added operational overhead for a three-person team supporting multiple brands. ## GitHub Copilot is now available in the AI SDK harness layer DevFeed: [GitHub Copilot is now available in the AI SDK harness layer]() Original publisher: [Read original article]() Author: Felix Arntz Published: 2026-09-10T17:39:00Z Content type: release Language: en Sources: [Vercel News]() Topics: [GitHub Copilot](), [vercel ai sdk](), [AI-assisted coding](), [cursor]() Tags: [agent](), [ai](), [claude](), [claude-code](), [code](), [codex](), [coding](), [copilot](), [cursor](), [github](), [github-copilot]() ### AI overview The AI SDK harness layer adds official GitHub Copilot support through an adapter, allowing applications to run Copilot and other coding agents through the same HarnessAgent interface. ### Source excerpt The AI SDK harness layer now supports GitHub Copilot through the official @ai-sdk/harness-github-copilot adapter. The harness layer lets your application run different coding agents through the same HarnessAgent interface, so you can switch agents without changing your application code. Pass githubCopilot to HarnessAgent: Under the hood, the adapter uses @ai-sdk/harness-acp to connect GitHub Copilot to HarnessAgent through the Agent Client Protocol (ACP). Supported harnesses now include, in addition to GitHub Copilot, Claude Code, Cline, Codex, Cursor, Deep Agents, fx, Grok Build, OpenCode, and Pi, with more coming soon. Read the GitHub Copilot harness documentation to get started. Read more ## Tako Search is free on AI Gateway through September 30 DevFeed: [Tako Search is free on AI Gateway through September 30]() Original publisher: [Read original article]() Author: Jerilyn Zheng Published: 2026-09-10T00:00:00Z Content type: release Language: en Sources: [Vercel News]() Topics: [vercel ai sdk](), [SDKs](), [API](), [Web](), [data]() Tags: [ai](), [ai-gateway](), [api](), [configuration](), [documentation](), [free](), [integration](), [playground](), [sdk](), [search](), [tools](), [web]() ### AI overview This changelog announces that Tako Search is free on AI Gateway through September 30. It enables AI models to search curated data and the live web, filter results, and provide current answers with citations and visualizations. Tako Search works with any model on AI Gateway and does not require a separate Tako account or API key. ### Source excerpt Tako Search is free exclusively on AI Gateway through September 30. It lets AI models search Tako's curated data and the live web, filter web results by domain or publication date, and use the results to answer questions with current information, citations, and visualizations. After September 30, searches are billed at standard rates. The same integration works with any model on AI Gateway, so you can switch models without changing your search setup. You also don't need a separate Tako account or API key. To use Tako Search with the AI SDK, add gateway.tools.takoSearch() to a generateText or streamText request. The model can then call it when it needs current information: Try Tako Search in the AI Gateway playground. See the web search documentation for configuration and search options. Read more ## fx is now available in the AI SDK harness layer DevFeed: [fx is now available in the AI SDK harness layer]() Original publisher: [Read original article]() Author: Felix Arntz Published: 2026-08-31T06:00:00Z Content type: release Language: en Sources: [Vercel News]() Topics: [vercel ai sdk](), [SDKs](), [Vercel](), [AI-assisted coding](), [API](), [Protocol (disambiguation)]() Tags: [agent](), [agents](), [ai](), [api](), [building](), [claude](), [claude-code](), [codex](), [coding-agents](), [cursor](), [integration](), [open-source](), [opencode](), [sdk](), [vercel]() ### AI overview Vercel's AI SDK harness layer now supports fx, a lightweight open-source coding agent. Developers can run fx in their applications through the HarnessAgent interface using the official @ai-sdk/harness-fx adapter, which connects through the Agent Client Protocol via @ai-sdk/harness-acp. ### Source excerpt The AI SDK harness layer now supports fx, Vercel's lightweight, open-source coding agent. The harness layer provides one API for running coding agents in your application, so you can add fx without building a separate integration. Configure HarnessAgent with the official @ai-sdk/harness-fx adapter: The @ai-sdk/harness-fx adapter connects to fx over the Agent Client Protocol (ACP) using @ai-sdk/harness-acp. fx joins Claude Code, Cline, Codex, Cursor, Deep Agents, Grok Build, OpenCode, and Pi in the list of supported harnesses. Read the fx harness documentation to get started. Read more ## CoreBreak proves agent guardrails need to live outside the agent DevFeed: [CoreBreak proves agent guardrails need to live outside the agent]() Original publisher: [Read original article]() Author: Tyler Akidau Published: 2026-08-31T00:00:00Z Content type: opinion Language: en Sources: [Redpanda]() Topics: [Artificial Intelligence](), [AI Agent](), [Security](), [Securing AI](), [Amazon Web Services](), [vercel ai sdk](), [Amazon Bedrock](), [Google](), [Vercel]() Tags: [agent](), [agentic-ai](), [ai](), [ai-agent](), [architecture](), [aws](), [bedrock](), [cve](), [google](), [policy](), [security](), [thought-leadership](), [vercel]() ### AI overview The article argues that CoreBreak exposed a structural security flaw in major AI agent stacks and that agent guardrails should be enforced outside the agent's control. ### Source excerpt At Black Hat 2026, CoreBreak exposed the same structural flaw across AWS, Google, and Vercel. Here's why AI agent security enforcement should live beyond the agent's reach. ## Hy4 Preview now available on AI Gateway DevFeed: [Hy4 Preview now available on AI Gateway]() Original publisher: [Read original article]() Author: Jerilyn Zheng Published: 2026-08-28T00:00:00Z Content type: release Language: en Sources: [Vercel News]() Topics: [Language models](), [vercel ai sdk](), [AI-assisted coding](), [AI, ML & Data Engineering]() Tags: [ai-gateway](), [api](), [coding](), [inference](), [language-models](), [performance]() ### AI overview Hy4 Preview, Tencent's open-source Mixture-of-Experts model, is now available through AI Gateway. The release describes AI SDK and coding-agent setup, along with gateway features for model access, usage tracking, routing, and pricing. ### Source excerpt Hy4 Preview from Tencent is now available on AI Gateway. Hy4 Preview is an open-source Mixture-of-Experts model with 770B total parameters and 49B active per token, aimed at long-horizon coding, document analysis, game development, and scientific reasoning. It serves a context window of 1M tokens. To use Hy4 Preview, set model to tencent/hy4-preview in the AI SDK: To use it in a coding agent, see the coding agents guide, then run vercel ai-gateway coding-agents setup to connect agents like Claude Code, Codex, OpenCode, Cursor, Pi, and more and select tencent/hy4-preview inside the agent. Try Hy4 Preview in the model playground. AI Gateway provides a unified API for calling models, tracking usage and cost, and configuring retries, failover, and performance optimizations for higher-than-provider uptime. It includes built-in custom reporting, Zero Data Retention support, budgets for API keys, routing rules, and more. AI Gateway reflects provider pricing with no markup and does not charge a platform fee on inference, including on Bring Your Own Key (BYOK) requests. You can view all language models available on AI Gateway. Read more ## Gemini 3.5 Transcribe now available on AI Gateway DevFeed: [Gemini 3.5 Transcribe now available on AI Gateway]() Original publisher: [Read original article]() Author: Jerilyn Zheng Published: 2026-08-26T00:00:00Z Content type: release Language: en Sources: [Vercel News]() Topics: [Artificial Intelligence](), [Google](), [Streaming](), [SDKs](), [vercel ai sdk](), [WebSocket]() Tags: [ai](), [ai-gateway](), [audio](), [gemini](), [google](), [sdk](), [speech](), [streaming](), [transcription]() ### AI overview Vercel announces that Google's Gemini 3.5 Transcribe models are available through AI Gateway for both complete recordings and live audio. The models support multilingual detection across more than 85 languages, custom vocabulary, speaker identification, word-level timestamps, and streaming transcription through AI SDK 7. ### Source excerpt Gemini 3.5 Transcribe from Google is now available on AI Gateway for recorded and live audio: google/gemini-3.5-transcribe transcribes a complete audio file in one request. google/gemini-3.5-transcribe-live transcribes audio over a WebSocket and returns text as the audio arrives. Both models automatically detect more than 85 languages, including when a speaker switches languages. You can also provide custom vocabulary to improve the transcription of names, technical terms, and uncommon spellings. The model for complete recordings can also identify speakers and return word-level timestamps. Streaming transcription is available in AI SDK 7. Install the latest AI SDK and AI Gateway provider: Transcribe live audio Use streamTranscribe with a ReadableStream of raw audio chunks. Set inputAudioFormat to match the audio being sent: Transcribe a complete recording Use transcribe to send a complete audio file and receive the finished transcript: Try Gemini 3.5 Transcribe Live in the model playground, browse all transcription models, or read the speech quickstart. Read more ## Muse Image now available on AI Gateway DevFeed: [Muse Image now available on AI Gateway]() Original publisher: [Read original article]() Author: Jerilyn Zheng Published: 2026-08-26T00:00:00Z Content type: release Language: en Sources: [Vercel News]() Topics: [Artificial Intelligence](), [Meta](), [vercel ai sdk](), [API](), [superintelligence](), [Inference]() Tags: [ai](), [ai-gateway](), [api](), [image](), [inference](), [meta](), [meta-muse](), [model](), [models](), [muse](), [pricing](), [sdk](), [superintelligence](), [uptime]() ### AI overview Muse Image, Meta Superintelligence Labs' first image model, is now available through Vercel's AI Gateway. It supports both image generation from prompts and instruction-based image editing, including reference images, through the AI SDK. AI Gateway also provides unified model access, usage and cost tracking, failover, routing, reporting, budgets, and performance optimizations without markup or platform fees on inference. ### Source excerpt Muse Image from Meta Superintelligence Labs is now available on AI Gateway. It is their first image model and a separate family from Muse Spark, returning images rather than text. Send a prompt and get an image back, or send an image with an instruction and get it changed. One model does both, so you don't switch models to move from generating to editing. To use Muse Image, set model to meta/muse-image-1.0 and call generateImage from the AI SDK: To steer the result toward art you already have, pass reference images in prompt.images alongside the text, and the model blends them into what it draws. Editing Pass the image you want changed in prompt.images with an instruction, and the model changes what you asked for and leaves the rest: Try Muse Image in the model playground. AI Gateway provides a unified API for calling models, tracking usage and cost, failover, and performance optimizations for higher-than-provider uptime. It includes built-in custom reporting, budgets for API keys, routing rules, and more. AI Gateway reflects provider pricing with no markup and does not charge a platform fee on inference, including on Bring Your Own Key (BYOK) requests. You can view all image models available on AI Gateway. Read more ## AI Gateway now supports asynchronous video generation DevFeed: [AI Gateway now supports asynchronous video generation]() Original publisher: [Read original article]() Author: Jerilyn Zheng Published: 2026-08-25T00:00:00Z Content type: release Language: en Sources: [Vercel News]() Topics: [vercel ai sdk](), [SDKs](), [web applications]() Tags: [ai](), [ai-gateway](), [event](), [generation](), [http](), [image-to-video](), [text-to-video](), [video](), [workflow]() ### AI overview Vercel AI Gateway adds asynchronous video generation through workflow webhooks, direct webhooks, polling, or later retrieval, avoiding long-lived requests that may time out. ### Source excerpt Video generation on AI Gateway can now run asynchronously. By default, generateVideo keeps one HTTP request to AI Gateway open until the result is ready. Because video generation can take seconds or minutes, that request can exceed request timeouts. With asynchronous generation, your application can receive a webhook, poll for completion, or start a generation and retrieve the result in a later request. Choose an option based on whether your process can keep running and whether your application can receive webhooks: Existing generateVideo calls continue to work as before. All four options support text-to-video, image-to-video, reference-to-video, and other video inputs. Upgrade the SDK Install the latest versions of the AI SDK and AI Gateway provider: Use asynchronous video generationWait for completion in a Workflow An easy way to consume the completion webhook is a Workflow SDK. The workflow creates its own webhook URL, passes it to startVideo, and suspends until AI Gateway delivers the completion event. Install the Workflow SDK alongside the AI SDK: While the video renders, the workflow run is suspended and resumes when AI Gateway delivers the terminal event. Use a webhook with generateVideo Pass webhook to generateVideo to wait for a completion event without polling. AI Gateway sends an event when the job completes or fails. The SDK waits for that event, fetches the generated videos, and resolves the original generateVideo call. The calling process and webhook handler need a shared token and store so the delivery can be matched to the correct generation. generateVideo does not expose the signing secret for this job. See the webhook verification documentation for the complete receiver pattern. Both the polling and webhook options for generateVideo return result.videos as GeneratedFile objects. The SDK downloads provider-hosted videos, making uint8Array, base64, and mediaType available in either job. Poll with generateVideo Add poll to an existing generateVideo ca ## Vercel Connect is now generally available DevFeed: [Vercel Connect is now generally available]() Original publisher: [Read original article]() Author: Dima Voytenko Published: 2026-08-25T00:00:00Z Content type: release Language: en Sources: [Vercel News]() Topics: [SDKs](), [vercel ai sdk]() Tags: [agent](), [agents](), [ai](), [auth](), [authentication](), [cli](), [deployment](), [mcp](), [oauth](), [observability](), [secrets](), [slack](), [vercel]() ### AI overview Vercel Connect is generally available, providing runtime-scoped tokens, OIDC-based deployment authentication, connector management, and access controls without storing long-lived provider secrets. ### Source excerpt Vercel Connect is now generally available on all plans and in v0. Instead of storing long-lived provider secrets, your code requests short-lived, scoped tokens at runtime. Deployments authenticate with their existing Vercel OIDC identity. Each token is scoped to the task, refreshed automatically, and expires on its own. Any service with one command Register a connector once from the CLI. Pass the service name and the CLI pre-populates the brand name, icon, auth type, and MCP or discovery URL, then prompts for any credentials the service requires: Connect ships with 100+ preset connectors for tools like Notion and Workday, managed connectors for Slack, GitHub, Linear, Salesforce, Snowflake, and Microsoft, plus generic OAuth, API key authentication, and MCP servers. Tokens at runtime, not secrets at rest Request a token only when your code needs one, with getToken: Switch the subject from the app to a named user and the token acts on that user's behalf, triggering the authorization flow when consent is needed. Access your team can inspect and prove New at GA: Fine-grained RBAC controls who can create and manage connectors Audit logs record authorization and connector activity Token and trigger observability shows how tokens are used across projects Together with per-environment attachment, including Custom Environments, and one-command revocation, external access becomes something your team can inspect, prove, and cut off in seconds. Triggers without stored webhook secrets Triggers deliver provider events to your app without a stored secret. Vercel Connect verifies signatures server-side, re-attests each event using an OIDC identity, and forwards the event to your project, even when Deployment Protection is enabled. Connect also manages private keys and the full credential lifecycle, so you get standards-compliant OAuth without building the infrastructure yourself. Works with your stack Connect works wherever your functions run and is supported in v0, eve, and Chat SD ## Cline is now available in the AI SDK harness layer DevFeed: [Cline is now available in the AI SDK harness layer]() Original publisher: [Read original article]() Author: Felix Arntz Published: 2026-08-18T00:00:00Z Content type: release Language: en Sources: [Vercel News]() Topics: [vercel ai sdk](), [SDKs](), [Code](), [Shell](), [Command-line interface]() Tags: [agent](), [agents](), [ai](), [claude-code](), [codex](), [filesystem](), [opencode](), [remote](), [sandbox](), [sdk]() ### AI overview Vercel's AI SDK harness layer now supports Cline through the official @ai-sdk/harness-cline adapter. Cline uses the shared HarnessAgent interface, runs in the host process, and accesses the sandbox as a remote filesystem and shell. ### Source excerpt The AI SDK harness layer lets you run established coding-agent runtimes through one unified interface, so you can switch runtimes without changing your application code. Today we are adding Cline, which runs through the same HarnessAgent interface as other supported harness. @ai-sdk/harness-cline is the official harness adapter for Cline, implemented in collaboration with the Cline team. Cline runs fully in the host process and uses the sandbox as a remote filesystem and shell, so only its tools operate in the sandbox. No bridge process is installed inside the sandbox, similar to the existing Pi harness. Read the Cline harness documentation to get started. The full supported list of harnesses is now: Claude Code, Cline, Codex, Deep Agents, Grok Build, OpenCode, and Pi, with more coming soon. Read more ## Gemini 3.7 Flash now available on AI Gateway for 50% off DevFeed: [Gemini 3.7 Flash now available on AI Gateway for 50% off]() Original publisher: [Read original article]() Author: Jerilyn Zheng Published: 2026-08-13T00:00:00Z Content type: release Language: en Sources: [Vercel News]() Topics: [Artificial Intelligence](), [AI Models](), [Google](), [vercel ai sdk](), [AI-assisted coding](), [Vercel](), [API](), [web applications]() Tags: [2026](), [agentic](), [ai](), [api](), [api-keys](), [claude](), [claude-code](), [code](), [codex](), [coding](), [gemini](), [google](), [inference](), [models](), [opencode](), [performance](), [playground](), [pricing](), [routing](), [sdk]() ### AI overview Vercel announces that Google's Gemini 3.7 Flash is available through AI Gateway at 50% off until December 31, 2026. The release highlights improved reliability for software engineering and agentic work, stronger adherence when generating application code from design mocks, coding-agent integrations, a model playground, and gateway features for usage, cost, routing, retries, failover, and uptime. ### Source excerpt Gemini 3.7 Flash from Google is now available on AI Gateway for 50% off till December 31st, 2026. Gemini 3.7 Flash improves on prior Flash models at software engineering and agentic work. It resolves issues more reliably and spends less time stuck in failed agent loops, which matters on long tool-calling sequences where one derailment costs the rest of the run. It also generates desktop and web application code directly from design mocks, with closer adherence to the source design. To use Gemini 3.7 Flash, set model to google/gemini-3.7-flash in the AI SDK: To use it in a coding agent, run vercel ai-gateway coding-agents setup to connect Claude Code, Codex, OpenCode, or Pi, then select google/gemini-3.7-flash inside the agent. To try Gemini 3.7 Flash with no code, try the model in the model playground. AI Gateway provides a unified API for calling models, tracking usage and cost, and configuring retries, failover, and performance optimizations for higher-than-provider uptime. It includes built-in custom reporting, Zero Data Retention support, budgets for API keys, routing rules, and more. AI Gateway reflects provider pricing with no markup and does not charge a platform fee on inference, including on Bring Your Own Key (BYOK) requests. Read more ## Use ACP-compatible harnesses with the AI SDK harness layer DevFeed: [Use ACP-compatible harnesses with the AI SDK harness layer]() Original publisher: [Read original article]() Author: Felix Arntz Published: 2026-08-13T00:00:00Z Content type: release Language: en Sources: [Vercel News]() Topics: [vercel ai sdk](), [Protocol (disambiguation)](), [SDKs](), [coding](), [Claude Code](), [codex]() Tags: [agent](), [ai](), [claude-code](), [codex](), [integration](), [opencode](), [sdk]() ### AI overview The AI SDK harness layer now supports Agent Client Protocol-compatible harnesses through the new @ai-sdk/harness-acp package. It acts as a meta adapter for harnesses that provide ACP-compatible packages, while direct adapters remain preferable for Claude Code and Codex when tighter integration is needed. ### Source excerpt The AI SDK harness layer now supports any Agent Client Protocol (ACP)-compatible harness with HarnessAgent through the new @ai-sdk/harness-acp package. Previously, every harness adapter wrapped one specific runtime (Claude Code, Codex, Pi, Deep Agents, OpenCode). @ai-sdk/harness-acp wraps the protocol instead. It is a meta adapter: rather than adapting a single harness, it lets you build an adapter for any harness that ships an ACP-compatible package. Implement a harness by passing that package to createACP and configure basic harness mapping. Then pass it to HarnessAgent like any other harness: ACP is an abstraction over coding harnesses, but the AI SDK harness layer stays deliberately decoupled from it. Not every harness supports ACP, and for some, ACP limits or changes how well they expose their internal behaviors, so a direct adapter can produce a tighter integration. For Claude Code and Codex, prefer the direct @ai-sdk/harness-claude-code and @ai-sdk/harness-codex adapters over an ACP-based implementation; reach for @ai-sdk/harness-acp when a harness has no direct adapter but ships an ACP-compatible package. Read the ACP harness documentation to get started. Read more ## Grok Build is now available in the AI SDK harness layer DevFeed: [Grok Build is now available in the AI SDK harness layer]() Original publisher: [Read original article]() Author: Maya Lekhi Published: 2026-08-13T00:00:00Z Content type: release Language: en Sources: [Vercel News]() Topics: [vercel ai sdk](), [SDKs](), [coding](), [Claude Code](), [codex]() Tags: [agent](), [agents](), [ai](), [claude-code](), [codex](), [documentation](), [opencode](), [sdk]() ### AI overview Vercel has added Grok Build to the AI SDK harness layer, allowing it to run through the same HarnessAgent interface as other supported coding-agent runtimes. The official Grok Build adapter is built on the ACP harness adapter. ### Source excerpt The AI SDK harness layer lets you run established coding-agent runtimes through one unified interface, so you can switch runtimes without changing your application code. Today we are adding Grok Build, which runs through the same HarnessAgent interface as every other supported harness. @ai-sdk/harness-grok-build is the official harness adapter for Grok Build, built on top of the ACP harness adapter (@ai-sdk/harness-acp). Read the Grok Build harness documentation to get started. The full supported list of harnesses is now: Claude Code, Codex, Deep Agents, Grok Build, OpenCode, Pi, with more coming soon. Read more ## Exa web search free through August 31 on AI Gateway and eve DevFeed: [Exa web search free through August 31 on AI Gateway and eve]() Original publisher: [Read original article]() Author: Shar Dara Published: 2026-08-12T00:01:00Z Content type: release Language: en Sources: [Vercel News]() Topics: [Web](), [vercel ai sdk](), [SDKs](), [API]() Tags: [ai](), [ai-gateway](), [api](), [documentation](), [sdk](), [search](), [web]() ### AI overview Exa web search is free on AI Gateway through August 31 and is the default web search for eve agents. It works with AI Gateway models without a separate Exa API key and provides web results and extracted page content with filtering and token-efficient excerpts. ### Source excerpt Exa web search is now free on AI Gateway through August 31, and it's now the default web search for eve agents. The tool works with any AI Gateway model, with no separate Exa API key. When the model calls it, AI Gateway routes the request to Exa's Search API, returning web results and extracted page content with support for domain filters, date filters, and token-efficient excerpts. The built-in web_search tool in eve now uses Exa by default when the model is served through AI Gateway. See other supported search configurations in the eve docs. To use it with the AI SDK, pass gateway.tools.exaSearch() to the tools parameter of a generateText call: Read the web search documentation to learn more and get started. Read more ## Grok 4.6 now available on AI Gateway DevFeed: [Grok 4.6 now available on AI Gateway]() Original publisher: [Read original article]() Author: Jerilyn Zheng Published: 2026-08-12T00:00:00Z Content type: release Language: en Sources: [Vercel News]() Topics: [AI, ML & Data Engineering](), [vercel ai sdk](), [long-context](), [AI-assisted coding]() Tags: [agent](), [ai](), [ai-gateway](), [ai-models](), [api](), [model](), [reasoning](), [vercel]() ### AI overview Grok 4.6 is now available through AI Gateway, with a 500K-token context window, text and image inputs, and configurable reasoning levels. The announcement also describes AI Gateway features for model access, usage and cost tracking, routing, reliability, and reporting. ### Source excerpt Grok 4.6 from SpaceXAI is now available on AI Gateway. The model has a 500K token context window and accepts text and image inputs. Grok 4.6 supports low, medium, high, and xhigh reasoning levels and defaults to high. To use Grok 4.6, set model to xai/grok-4.6 in the AI SDK: To use it in a coding agent, run vercel ai-gateway coding-agents setup to connect Claude Code, Codex, OpenCode, or Pi, then select xai/grok-4.6 inside the agent. To try Grok 4.5 with no code, try the model in the model playground. AI Gateway provides a unified API for calling models, tracking usage and cost, and configuring retries, failover, and performance optimizations for higher-than-provider uptime. It includes built-in custom reporting, Zero Data Retention support, budgets for API keys, routing rules, and more. AI Gateway reflects provider pricing with no markup and does not charge a platform fee on inference, including on Bring Your Own Key (BYOK) requests. Read more ## Grok Imagine Image 2.0 now available on Vercel AI Gateway DevFeed: [Grok Imagine Image 2.0 now available on Vercel AI Gateway]() Original publisher: [Read original article]() Author: Jerilyn Zheng Published: 2026-08-08T00:00:00Z Content type: release Language: en Sources: [Vercel News]() Topics: [Artificial Intelligence](), [Vercel](), [vercel ai sdk](), [SDKs]() Tags: [ai](), [images](), [model](), [models](), [sdk](), [vercel]() ### AI overview Grok Imagine Image 2.0 Preview from xAI is now available through Vercel AI Gateway. It supports detailed image generation, layout-aware typography, and image editing that preserves unchanged subjects and details. The article also explains how to call the model with the AI SDK, select 1k or 2k resolution, generate multiple images, and provide an input image for editing. ### Source excerpt Grok Imagine Image 2.0 Preview from xAI is now available on AI Gateway. The model follows detailed instructions closely and plans typography and layout together, so dense, multi-part visuals like infographics, posters, and title screens hold their structure and small text stays legible. Grok Imagine Image 2.0 Preview also supports image editing, keeping subjects and details consistent across repeated generations. Try the model out now in imagine.vercel.sh, running on AI Gateway. To use Grok Imagine Image 2.0 Preview, set model to xai/grok-imagine-image-2.0-preview and call generateImage from the AI SDK. Set resolution to 1k or 2k under providerOptions.xai to pick an output tier, and n for more than one image per call. For image editing, pass an image in prompt.images alongside the instruction, and the model changes what you asked for and leaves the rest: To view all image models supported on AI Gateway, see the full list here. Read more ## Seedance 2.5 now available on Vercel AI Gateway DevFeed: [Seedance 2.5 now available on Vercel AI Gateway]() Original publisher: [Read original article]() Author: Jerilyn Zheng Published: 2026-08-06T00:00:00Z Content type: release Language: en Sources: [Vercel News]() Topics: [Vercel](), [Artificial Intelligence](), [vercel ai sdk](), [SDKs](), [API]() Tags: [ai](), [ai-gateway](), [api](), [audio](), [generate](), [image](), [model](), [sdk](), [vercel](), [video]() ### AI overview Vercel AI Gateway now offers ByteDance's Seedance 2.5 video model. The model generates clips up to 30 seconds with synchronized audio, supports up to 50 multimodal reference inputs, can extend short clips, and can edit videos while preserving framing, camera movement, and pacing. The article also documents video generation, image and video references, and video editing through the AI SDK. ### Source excerpt Seedance 2.5 from ByteDance is now available on AI Gateway. It can generate up to 30-second clips with synchronized audio, and takes up to 50 reference inputs in one request, spanning images, video, audio, and style. A single clip holds camera movement and continuity without stitching shots together in post, and short clips can be extended, carrying over character, scene, and camera movement. Prompts work in more than ten languages. Seedance 2.5 can also edit a finished video in place, swapping backgrounds, products, or characters while the frame, camera, and pacing stay fixed, so one shoot yields several variants. Generating a video Set model to bytedance/seedance-2.5 and call generateVideo from the AI SDK: Adding image references Pass image URLs in inputReferences and point at them in the prompt with [Image 1], [Image 2]. Tag each entry with its media type. Seedance treats an untagged URL as an image. Combining reference types Mix videos and images in the same request to pull different qualities from each: motion and camera work from a clip, a subject's appearance from a still. Videos go in the same inputReferences array as images, tagged with a video media type. Numbering runs per type so the first video is [Video 1] and the first image is [Image 1], and both can appear in one prompt. Editing a video Referencing and editing use the same setup, so the prompt decides which you get. Ask the model to follow a clip's camera movement and it builds a new video from scratch. Ask it to change something in the clip and it edits the video you gave it. You can reference and edit in one prompt, though results get less predictable, so it helps to be explicit about what each asset is for. For the playground and API reference, see the model page, or browse every video model on AI Gateway. Read more ## Ling 3.0 Tiny is now available on AI Gateway DevFeed: [Ling 3.0 Tiny is now available on AI Gateway]() Original publisher: [Read original article]() Author: Jerilyn Zheng Published: 2026-08-06T00:00:00Z Content type: release Language: en Sources: [Vercel News]() Topics: [Artificial Intelligence](), [moe](), [vercel ai sdk](), [long-context](), [SDKs]() Tags: [ai](), [ai-gateway](), [api](), [caching](), [coding-agents](), [context-window](), [cost](), [models](), [moe](), [sdk]() ### AI overview Ling 3.0 Tiny from ANT Group is available on Vercel AI Gateway, initially in a free slot. It is a Mixture-of-Experts model with 7.9B total parameters, approximately 1.3B active parameters per token, a 256K-token context window, and support for up to 32K output tokens. The article also explains how to use it through the AI SDK, the model playground, and coding agents. ### Source excerpt Ling 3.0 Tiny from ANT Group is now on AI Gateway, free to use till 8:00am PT on 8/14. Ling 3.0 Tiny takes the free slot from Ling 3.0 Flash. Ling 3.0 Tiny is a MOE model with 7.9B total parameters and about 1.3B active per token, a 256K token context window, and up to 32K output tokens. The model is built for responsive agents, instruction following, and multi-turn conversation, with native function calling and prompt caching. To use Ling 3.0 Tiny, set model to inclusionai/ling-3.0-tiny-free in the AI SDK. On August 14th, the new model name will be inclusionai/ling-3.0-tiny. Try Ling 3.0 Tiny in the model playground. To use it in a coding agent, run vercel ai-gateway coding-agents setup and select inclusionai/ling-3.0-tiny-free inside the agent. AI Gateway provides a unified API for calling models, tracking usage and cost, and configuring retries, failover, and performance optimizations for higher-than-provider uptime. It includes built-in custom reporting, Zero Data Retention support, budgets for API keys, routing rules, and more. AI Gateway reflects provider pricing with no markup and does not charge a platform fee on inference, including on Bring Your Own Key (BYOK) requests. Read more ## Muse Spark 1.2 is now available on Vercel AI Gateway DevFeed: [Muse Spark 1.2 is now available on Vercel AI Gateway]() Original publisher: [Read original article]() Author: Jerilyn Zheng Published: 2026-08-05T00:00:00Z Content type: release Language: en Sources: [Vercel News]() Topics: [AI-assisted coding](), [vercel ai sdk](), [debugging]() Tags: [agent](), [ai-gateway](), [api](), [code-generation](), [debugging](), [muse](), [spark](), [vercel]() ### AI overview Muse Spark 1.2 is available on Vercel AI Gateway as a coding-focused model update, with stated improvements in code generation, debugging, codebase understanding, and iterative developer workflows. ### Source excerpt Muse Spark 1.2 from Meta is now available on AI Gateway. It is a coding-focused update to the previous Muse Spark model. While keeping its general capabilities, 1.2 ships with improvements in code generation, complex debugging, codebase understanding, and end-to-end developer workflows. The model is built for long-horizon work like generating whole repositories, building out large projects end to end, and sustaining iterative loops where it writes, compiles, profiles, and improves code over many rounds. To use Muse Spark 1.2, set model to meta/muse-spark-1.2 in the AI SDK: To use Muse Spark in a coding agent, run vercel ai-gateway coding-agents setup to connect Claude Code, Codex, OpenCode, or Pi, then select meta/muse-spark-1.2 inside the agent. AI Gateway provides a unified API for calling models, tracking usage and cost, and configuring retries, failover, and performance optimizations for higher-than-provider uptime. It includes built-in custom reporting, Zero Data Retention support, budgets for API keys, routing rules, and more. AI Gateway reflects provider pricing with no markup and does not charge a platform fee on inference, including on Bring Your Own Key (BYOK) requests. Try Muse Spark 1.2 in the model playground. Read more ## DeepSeek V4 Flash now runs updated weights on AI Gateway DevFeed: [DeepSeek V4 Flash now runs updated weights on AI Gateway]() Original publisher: [Read original article]() Author: Jerilyn Zheng Published: 2026-07-31T07:00:00Z Content type: release Language: en Sources: [Vercel News]() Topics: [AI, ML & Data Engineering](), [vercel ai sdk](), [AI-assisted coding]() Tags: [agent](), [agentic](), [ai-gateway](), [coding](), [deepseek](), [flash](), [inference](), [model](), [vercel]() ### AI overview DeepSeek V4 Flash now uses updated weights by default through AI Gateway, improving its Terminal-Bench score without requiring a model ID or code change. ### Source excerpt DeepSeek V4 Flash now runs on updated weights by default on AI Gateway, with notably stronger agentic capabilities. On Terminal-Bench, it scores 82.7, up 25.8 points from 56.9 in the April preview. Requests to deepseek/deepseek-v4-flash pick up the new weights automatically, with no change to the model ID or your code. For now, DeepSeek is the only provider serving the updated weights. Other providers, including ones with Zero Data Retention, are coming next week. To use the updated DeepSeek V4 Flash, set model to deepseek/deepseek-v4-flash in the AI SDK. AI Gateway will route to providers with the new weights by default: To run V4 Flash in a coding agent, use vercel ai-gateway coding-agents setup to connect your agents to AI Gateway, then select deepseek/deepseek-v4-flash in the agent's model configuration. See the coding agents guide. AI Gateway reflects provider pricing with no markup and does not charge a platform fee on inference, including on Bring Your Own Key (BYOK) requests. Learn more about AI Gateway, view the AI Gateway model leaderboard or try it in our model playground. Read more [Next page]()